Infection of human cells with HIV requires a receptor complex consisting of CD4 and one of several chemokine receptors. Early after strains to date have been show to be specific for the chemokine receptors CCR5 and/or CXCR4. Early after acquisition and during the asymptomatic phase, most viruses are tropic for CCR5, while late in the disease many viruses are specific for CXCR4 or both of these receptors. Individuals with a homozygous null mutation in the CCR5 gene are relatively resistant to infection with HIV, and their immune systems appear normal. Strategies aimed at interfering with the ability of CCR5 to function as an HIV receptor are hence likely to result in a significant decrease in viral load without impairing the immune system, The effects of blocking CXCR4 function in vivo are not yet known, but blocking HIV usage of this receptor may be particularly important for reversing or delaying the onset of immunodeficiency. Interference with HIV interactions with chemokine receptors can potentially be achieved with compounds that specifically bind to CCR5 and CXCR4 and block their ability to serve as viral receptors. Alternatively, blocking expression of CCR5 or CXCR4 by lymphocytes, monocytes, and macrophages may also be an effective means of interfering with infection. We propose to develop approaches to interfere with the expression of these molecules in vivo. Modified chemokines, particularly MIP-1beta, Rantes, and SDF-1, will be designed so that they are retained in the endoplasmic reticulum and thus block cell surface expression of the chemokine receptors. The effectiveness of these molecules will be tested in transgenic mice that express human CCR5 and CXCR4, both by introduction of the molecules as transgenes and by retroviral vector transduction. In additional collaborative studies, these approaches will be compared to ribozyme-directed inhibition of CCR5 expression. The role of CXCR4 in mice will be determined by gene targeting and particular attention will be devoted to effects on T helper cells and macrophages, the principal targets of HIV infection. In collaboration with Project 3, retroviral vectors designed to express the modified chemokines in macrophages and T helper cells will be used to transduce these genes into bone marrow cells of the CCR5/CXCR4 transgenic mice, and the effect on expression of both mouse and transgenic human receptors will be assessed. The immune functions of these animals will be studied in collaborations with the Immunology Core. Successful inhibition of expression of human CCR5 and CXCR4 in these transgenic mice, in the absence of significant effects on immune responses, all allow us to apply this technique to patients infected with strains of HIV-1 of known receptor phenotype.